-
Notifications
You must be signed in to change notification settings - Fork 55
Description
There doesn't seem to be any compression in sendme. Take, for example, a 1G file of zeroes:
$ dd if=/dev/zero of=empty bs=1G count=1
1+0 records in
1+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.526697 s, 2.0 GB/s
$ sendme send empty
⠁ [00:00:00] waiting for requests
⠒ [00:00:00] waiting for requests
imported file empty, 1.00 GiB, hash 14fd9525cba83ddee5526775575b1680a509bc95eaf47b4b7814c036365e4cc9
If I receive this file on another machine, it takes the entire gigabyte to send. I can, of course, compress it myself:
$ gzip empty
$ sendme send empty.gz
...in which case it'll receive it much faster. However, fiddling with compression tools is a lot of added friction on receiving a file, and file transfer tools like sendme are (in my opinion) designed to solve friction.
It would be nice if sendme has optional compression (and if there are multiple compression algorithms, the ability to pick one, e.g. --compression zlib
). Ideally it would let the user know when it's compressing a file, because it can take a long time for some files, and having no user feedback can make the user think it's stuck.
If compression is implemented, you should be able to turn it off (e.g. --compression none
). A lot of file types are already compressed in some way, and extra compression will waste my time and CPU more than it would pay off.
Metadata
Metadata
Assignees
Labels
Type
Projects
Status